PDA

View Full Version : Paul DePodesta's Philosophy



RedsManRick
11-15-2010, 03:00 PM
I just came across this article via HardballTimes. It really does a great job at capturing the sabermetric perspective in a nutshell.

http://web.archive.org/web/20040206175549/http://www.csfb.com/thoughtleaderforum/2003/depodesta_sidecolumn.shtml

What I'd give to have a better understanding of how Jocketty works or what he believes and how it differs. Obviously the devil is in the details here (how did the A's transformation lead to the Big 3?), but it's fascinating stuff.

bucksfan2
11-15-2010, 03:51 PM
I just came across this article via HardballTimes. It really does a great job at capturing the sabermetric perspective in a nutshell.

http://web.archive.org/web/20040206175549/http://www.csfb.com/thoughtleaderforum/2003/depodesta_sidecolumn.shtml

What I'd give to have a better understanding of how Jocketty works or what he believes and how it differs. Obviously the devil is in the details here (how did the A's transformation lead to the Big 3?), but it's fascinating stuff.

Interesting perspective. The A's would be a interesting study in FO methods. You have a numbers first front office that had a great, small market, run with the so called Big 3 but also PED infused offensive production.

I did love the blackjack reference. Hitting on 17 is a bad, bad idea..... but what if the player had a card count that was low and the dealer was showing a 10. IIRC in the book 21 they had a guy at the table playing wildly, maybe DePodesta was playing next to a guy amidst a card counting scheme.

RedsManRick
11-15-2010, 06:25 PM
I did love the blackjack reference. Hitting on 17 is a bad, bad idea..... but what if the player had a card count that was low and the dealer was showing a 10. IIRC in the book 21 they had a guy at the table playing wildly, maybe DePodesta was playing next to a guy amidst a card counting scheme.

I think that sort of exception thinking is probably valid on rare occasion, but too often it's used as justification for what's simply a bad decision -- like ignoring the LOOGY's massive platoon split because the LH you're sending up to bat is 3-5 with a homer against him. Sure, it's possible you found the exception to the rule -- but probably not. Sure, that 1 run you scored in the 1st inning at the expense of two outs made on purpose might be the game winner -- but probably not. Maybe Willy Taveras will start hitting the ball on the ground instead of swinging from his heels when we tell him to do it -- but probably not.

But you only need a few of those exceptions to pan out for you before you start thinking you're a genius who can beat the system. We all want to be the exception. But that line of thinking usually doesn't end well.

I love the part about thinking in probabilities. I think all teams do this to a point. But few go through the process of actually putting those probabilities on paper, doing the math and going with the decision suggested by it. From what I've read, in some front offices, that kind of analysis is only given heed when it confirmation the prevailing though, but ignored otherwise.

bucksfan2
11-16-2010, 09:28 AM
I think that sort of exception thinking is probably valid on rare occasion, but too often it's used as justification for what's simply a bad decision -- like ignoring the LOOGY's massive platoon split because the LH you're sending up to bat is 3-5 with a homer against him. Sure, it's possible you found the exception to the rule -- but probably not. Sure, that 1 run you scored in the 1st inning at the expense of two outs made on purpose might be the game winner -- but probably not. Maybe Willy Taveras will start hitting the ball on the ground instead of swinging from his heels when we tell him to do it -- but probably not.

But you only need a few of those exceptions to pan out for you before you start thinking you're a genius who can beat the system. We all want to be the exception. But that line of thinking usually doesn't end well.

I love the part about thinking in probabilities. I think all teams do this to a point. But few go through the process of actually putting those probabilities on paper, doing the math and going with the decision suggested by it. From what I've read, in some front offices, that kind of analysis is only given heed when it confirmation the prevailing though, but ignored otherwise.

I think you could make a valid argument that if you make it to the big leagues or a MLB front office you are an exception. I think a coach once told me that 1% of knothole baseball players make it to college baseball and then 1% of those players make it to professional baseball. So in essence you are dealing with exceptions when running a baseball team.

As for your LOGGY example I think it swings both ways. You have a valid point that they guy has had a great deal of success against the particular pitcher. He is 3-5 with a HR. It shouldn't happen based upon splits but it has happened so far. It could be as simple as the hitters batting stance allows the hitter to pick up the ball early and make a good swing on the ball. If you see the ball well it makes hitting it much easier. On the other hand it could be dumb luck. Guy could have put 3 good swings on the ball and thats it. Over the course of the next 20 AB's between the to the guy may go 1-20.

Johnny Footstool
11-16-2010, 11:40 AM
As for your LOGGY example I think it swings both ways. You have a valid point that they guy has had a great deal of success against the particular pitcher. He is 3-5 with a HR. It shouldn't happen based upon splits but it has happened so far. It could be as simple as the hitters batting stance allows the hitter to pick up the ball early and make a good swing on the ball. If you see the ball well it makes hitting it much easier. On the other hand it could be dumb luck. Guy could have put 3 good swings on the ball and thats it. Over the course of the next 20 AB's between the to the guy may go 1-20.

This is where scouts can be the most valuable. In those previous 5 ABs, was the pitcher controlling the AB until making a mistake? Was the hitter sitting on the fastball and whaling on the first one he saw? There are a ton of questions that have to be answered by observation and analysis. The numbers tell you that something is happening, but you've still got to figure out what.

I think that every baseball team out there knows this. I seriously doubt that anyone simply runs their team like a Strat-O-Matic game.

westofyou
11-16-2010, 11:51 AM
I think that every baseball team out there knows this. I seriously doubt that anyone simply runs their team like a Strat-O-Matic game.

Yep, because a strat game can't tell you that the LOOGY has a slider that runs in on hitters and that hitter A happens to feast on that sort of pitch while hitters B-Z can't handle it one iota.

IslandRed
11-17-2010, 02:53 AM
Reading that article -- again? It looked familiar -- along with the stuff about Paul in Moneyball makes me believe that his best role is as an assistant GM. The head GM needs to balance both sources of information, the stats and the scouts, because both are important. Anyone who's dismissive of one side or the other is going to struggle, and DePodesta's contempt for the scouting side is evident. (Contrast his experience with Bill James, who, after working inside the game for awhile, was willing to concede that maybe some people in the game knew something about baseball.)

But like Johnny said, no team's totally one or the other anymore.

As for the blackjack comparisons... Blackjack is just percentages and randomness. You're not playing against anyone, just the odds. To me, baseball is more like poker. The probabilities are important but don't exist in a vacuum, because there's an opponent across the way whose actions affect your probabilities. Specifically, if you always play the percentages, you become predictable to anyone whose information is sufficiently good, and thus are susceptible to probability-reducing countermeasures.

bucksfan2
11-17-2010, 09:56 AM
Reading that article -- again? It looked familiar -- along with the stuff about Paul in Moneyball makes me believe that his best role is as an assistant GM. The head GM needs to balance both sources of information, the stats and the scouts, because both are important. Anyone who's dismissive of one side or the other is going to struggle, and DePodesta's contempt for the scouting side is evident. (Contrast his experience with Bill James, who, after working inside the game for awhile, was willing to concede that maybe some people in the game knew something about baseball.)

But like Johnny said, no team's totally one or the other anymore.

As for the blackjack comparisons... Blackjack is just percentages and randomness. You're not playing against anyone, just the odds. To me, baseball is more like poker. The probabilities are important but don't exist in a vacuum, because there's an opponent across the way whose actions affect your probabilities. Specifically, if you always play the percentages, you become predictable to anyone whose information is sufficiently good, and thus are susceptible to probability-reducing countermeasures.

I liked the black jack comparisons but in a different way. Hitting on 17 isn't really a good idea in blackjack. Although there isn't much of a difference between hitting on 16 and 17 but I digress. I brought it up earlier in this thread but will bring it up again. In the book 21 the black jack scheme worked like this. You had your basic player who played by the book and bet the table minimum. This player was to keep the count and signal in the big better when the count got favorable. The big better would play similar to the basic player just by putting a lot of money on the table. Then they had a wildcard. The wild card would do exactly that, play wild. Raise and lower their best for no apparent reason, double down on hard 12's, hit on 17, stay on 12 into a 20, etc. They played like an idiot, but an idiot with a large bankroll. The key to the wild card was not only to win (they played when the count was in their favor) but also to draw attention away from your big player who is raking it in.

I guess I liken it to this. While DePodesta is mad at the player to his right who is hitting on 17 and getting a 4 he is missing the guy to his left that is raking in the money. It kind of reminds me of Jocketty. While everyone is up in arms about his lack of movement, bad signings, or trades, he silently is building a competitive ball club. While he was bringing in guys like Cairo, Nix, Gomes, etc. people were complaining about the lack of overall improvement while behind the scenes he was working on bringing the most high profile Cuban defector in decades.

RedsManRick
11-17-2010, 11:40 AM
I guess I liken it to this. While DePodesta is mad at the player to his right who is hitting on 17 and getting a 4 he is missing the guy to his left that is raking in the money. It kind of reminds me of Jocketty. While everyone is up in arms about his lack of movement, bad signings, or trades, he silently is building a competitive ball club. While he was bringing in guys like Cairo, Nix, Gomes, etc. people were complaining about the lack of overall improvement while behind the scenes he was working on bringing the most high profile Cuban defector in decades.

You are suggesting that Jocketty has some insight that Depo doesn't and that Depo. He would disagree.

He would call Jocketty, a guy who generally makes sound decisions, but who doesn't avail himself of all of the available information (heavily leaning on scouting information, eschew advanced statistical analysis -- Jocketty is sitting in for an archetype here, this may not be a fully accurate description of him). He'll make pretty much the same calls as anybody else, won't put his stack at risk, and comes out ahead, but not by much.

GMs like Dave Littlefield and Ed Wade are the guys hitting on 17, going by some combination of their gut and improperly applied analysis. Occasionally these guys get the 4 and win big. But they lose in the long run.

Depo sees himself as the guy who surveys the room looking for a hot table (e.g. market inefficiencies). Once there, he then plays a straightforward, dispassionate game based on the odds.

Now, it's a gross simplification to suggest that non-saber GMs are the same level of irresponsible as the wild player at the table. And blackjack isn't a perfect analogy because it's risk-reward model is symmetric whereas baseball's isn't. (There's no equivalent to drafting Albert Pujols in blackjack)

I think we can agree that all GMs make decisions based on maximizing their odds for success. That's not really the point of differentiation. Rather, the question is how those odds are assessed. Are they done so in via a transparent, intellectually rigorous manner? Don't get me wrong, by turning away good information from scouts, Depo may not be living up to the sabermetric ideal. Sabermetrics is not about stats. It's about systematically analyzing all of the available information, qualitative and quantitative to come up with the best decisions -- valuing the process and it's results over the person who conducts it.